[Original developments of navigation techniques in traumatology]. / Développements originaux des techniques de navigation en traumatologie.

This article presents an original method for long objects modeling and two navigation applications in trauma surgery. Both concern long bone fracture treatment. Our modeling method requires only two x-ray views. The projection cones of the object are determined and their intersection is computed, providing an approached 30 model, which can be improved by adding a priori knowledge or other information. The first application concerns the control of diaphyseal fracture reduction treated by external fixation. Reference frames are fixed to the bone fragments and tracked by a 3D optical localizer, allowing the computation of their relative position. Approached 3D models of the fragments are displayed in real time according to the manipulation effected by the surgeon. The principal axes of the fragments, very useful for the fracture reduction, are also displayed. The alignment of the bone fragments is quantified by parameters provided in real time during the reduction. The second application concerns the distal locking of intra-medullary nails. A 3D model of the nail and its locking holes is built from two calibrated fluoroscopic views. The nail and the surgical tool are tracked thanks to reference frames fixed to each of them. A 3D view of these two elements is displayed in real time, guiding the surgeon in the difficult task of distal targeting. Experiments and results are presented for both applications. These techniques provide real 3D models to the surgeon during the operation, allowing precise guidance of the surgical gesture and considerable reduction of the irradiation to the patient and the surgical team.

Factors influencing the spatial precision of electromagnetic tracking systems used for MEG/EEG source imaging.

OBJECTIVES: The purpose of this study is to determine the factors influencing the spatial precision and the replicability of electromagnetic trackers (EMT) for the localization of electrodes and natural landmarks on the patient's head. MATERIALS AND METHODS: The effects of seven conditions on the measurement of the EMT were investigated with a Polhemus Fastrack: distance, contact between two components of the EMT, presence of magnetic object, localization of landmarks and electrodes on a phantom and a human subject without and with movements. RESULTS: The EMT has a precision of 0.15mm+/-0.36mm for the measurements made on still objects in a non-magnetic environment. On a human subject, the mean variation of the nasion position is 1.6mm+/-1.46mm and 2.7mm+/-1.40mm for the tragus. The increase of the electrode measurement dispersions is significant between the phantom and the human subject with a mean variation of 2.39mm+/-1.26mm. In certain conditions, up to 15% of the measurements may be considered as outliers. CONCLUSION: The precision significantly decreases for this application in the following cases: (1) physical contacts between the stylus/transmitter/receiver cables, (2) presence of magnetic objects in the surrounding of the EMT system, (3) skin and hair softness and (4) subject's head movements.

Conception of a navigation system controlling diaphyseal fracture reduction treated with external fixation.

BACKGROUND: The reduction of long bone fractures treated with external fixation is usually performed with fluoroscopic images, which include several disadvantages: 2D information, distortions, and irradiation to the patient and the surgical team. This article presents a new navigation technique to control the reduction of such fractures while minimizing the irradiation. METHODS: Optically tracked markers are fixed to pins inserted into the bone fragments. These last are modelled using two initial calibrated radiographs. The models can be improved with several types of anatomical data and are displayed in real time. RESULTS: This navigation system was tested on dry bones and an anatomical specimen leg. CONCLUSIONS: This new technique allows the visualization of the fracture in real time and from any viewpoint during the reduction. Irradiation is minimized using only two X-ray images.